The architecture and design process of a highly autonomous, situated agent is described. The architecture consists of a population of agents which are designed in an incremental, bottom-up fashion according to what we call a behavioural engineering approach. The agents operate in parallel and have a close coupling between perception and action. Emergent behaviour and memory-based learning deter...

Grammatical relationships (Glls) form an impor tant level of natural language processing, but different sets of ORs are useflfl for different purposes. Theretbre, one may often only have time to obtain a small training corpus with the desired GI1. annotations. On s u & a small training corpus, we compare two systems. They use difl'erent learning tedmiques, but we find that this difference by it...

We describe results on pitch accent placement in Dutch text obtained with a memory-based learning approach. The training material consists of newspaper texts that have been prosodically annotated by humans, and subsequently enriched with linguistic features and informational metrics using generally available, lowcost, shallow, knowledge-poor tools. We report on the effects of context-modelling ...

Traditional economic theory often describes real-world social dilemmas as abstract games where an individual’s goal is to maximize economic benefit by cooperating or competing with others. Despite extensive empirical work, descriptive models of human behavior in social dilemmas are lacking in both cognitive realism and predictive power. This article addresses a central challenge arising from th...

Scope classiication is a new instance-based learning (IBL) technique with a rule-based characterisation. Within the scope approach, the classiication of an object o is based on the examples that are closer to o than every example labelled with another class. In contrast to standard distance-based IBL classiiers, scope classiication relies on partial pre-orderings o between examples, indexed by ...

We give an overview of a new set of learningbased algorithms for fast and robust motion planning with noisy sensor data. These include improved search in configuration spaces using instance-based learning, and robust proximity queries in configuration spaces with noisy data. We demonstrate the performance of these new proximity and planning algorithms on the PR2 robot with noisy point-cloud data.

We evaluate the effect of automatically generated semantic clusters as as information source in our machine learning approach to the task of coreference resolution for Dutch. We compare these clusters which group semantically similar nouns together, to two semantic features based on WordNet encoding synonym and hypernym relations between nouns. Our experiments with two learners show that the cl...

This paper describes the development of a memory-based lemmatiser for Afrikaans called Lia. The paper commences with a brief overview of Afrikaans lemmatisation and it is indicated that lemmatisation is seen as a simplified process of morphological analysis within the context of this paper. This overview is followed by an introduction to memory-based learning – the machine learning technique th...

Case-based approaches to classification, as instance-based learning techniques, have a particular reliance on training examples that other supervised learning techniques do not have. In this paper we present the RDCL case profiling technique that categorises each case in a casebase based on its classification by the case-base, the benefit it has and/or the damage it causes by its inclusion in t...

In this paper, we consider the problem of leveraging existing fully labeled categories to improve the weakly supervised detection (WSD) of new object categories, which we refer to as mixed supervised detection (MSD). Different from previous MSD methods that directly transfer the pre-trained object detectors from existing categories to new categories, we propose a more reasonable and robust obje...